Loading...
Please wait, while we are loading the content...
Similar Documents
A Sequential Monte Carlo Sampling Approach for Cell Population Deconvolution from Microarray Data
| Content Provider | Semantic Scholar |
|---|---|
| Author | Roy, Sushmita Lane, Terran Allen, Chris Aragon, Anthony D. Werner-Washburne, Margaret |
| Copyright Year | 2004 |
| Abstract | Microarray analyses assume that gene expressions are measured from synchronous, homogeneous cell populations. In reality, the measured gene expression is produced by a mixture of populations in different stages of the cellular life cycle. Hence, it is important to estimate the proportions of cells in different stages and to incorporate this information in the analysis of gene expression, clusters, or function. In this paper we propose a novel unsupervised learning approach that models biological processes, such as the cell-cycle, as a hidden state-space model and uses a particle-filter based approach for estimating model parameters. Our approach finds a maximum a posteriori (MAP) estimate of the cell proportions given the gene expression and an estimate of the stage dependant gene expression. Evaluation of statistical validity of our approach using randomized data tests reveals that our model captures true temporal dynamics of the data. We have applied our approach to model the yeast cell-cycle and extracted profiles of the population dynamics for different stages of the cell-cycle. Our results are in agreement with biological knowledge and reproducible in multiple runs of our algorithm, suggesting that our approach is capable of extracting biologically meaningful and statistically significant information. Finally, the stage dependant gene expression can be used to determine clusters of active genes on a per-stage basis. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.cs.unm.edu/~treport/tr/04-11/recomb.pdf |
| Alternate Webpage(s) | https://www.cs.unm.edu/~treport/tr/04-11/recomb.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
